Let your data flow

Introducing Power BI dataflows

Wolfgang Strasser, MVP

Let your data flow – Introducing Power BI dataflows

Power BI serves as a self-service BI platform with a strong focus on data preparation and interactive analysis. With the introduction of Power BI dataflows, self-service data preparation is brought to a new level. The main concepts used are:

• Usage of common and mature technologies: data is stored as entities following the Common Data Model in Azure Data Lake Storage Gen2;

• Integration: dataflows are created and managed in Power BI app workspace

• Self-Service and low-code/no-code - Power Query is used as data preparation engine

• Connectivity: dataflows will support a variety of different data sources (including cloud-based and on-premises sources)

Join this session if you would like to learn more about the basic concepts and especially see Power BI dataflows in action.

WHY Power BI dataflows?

Business Intelligence needs

DATAbut: Data Preparation takes time

(60-80% of a BI project)

Self-Service BI needs

More “Power” for the key users / business analysts

Faster access to new data sources

Integration

Data Models

Self-Service BI – Current State

BIETL

✓

✓

Screenshot taken from: https://myignite.techcommunity.microsoft.com/sessions/65437

Power BI Report 3

Dataset

Power Query (M)

Power BI Report 2

Dataset

Power Query (M)

Power BI Report 1

ETL in the Context of Power BI

Dataset

Power Query (M)

Introducing Power BI dataflows

Power BI information hierarchy

Source: Amir Netz: Power BI dataflows Whitepaper (https://go.microsoft.com/fwlink/?linkid=2034388&clcid=0x409)

Good to Know

App workspaces only (Not available in “My workspace”)

Entities ~ Tables

Dataflow = data preparation pipeline

Power Query / M

Multiple dataflows

Transaction Consistency

Dataflow Calculation engine

Good to Know

Re-usable data integration

Created / managed in the cloud

Proven technologies

Power Query

Azure Data Lake Storage Gen2

Common Data Model

Common Data Model

Data SchemaStandardized, modular, extensible

Consists ofEntities

Attributes

Semantic metadata

relationships

Taken from https://aka.ms/cdmposterhttps://aka.ms/cdmrepo CDM GitHub Repo

CDS for Apps Power BI

Data IntegrationCommon Data Model

ADLS v2Enrichment and AI

IntelligenceIntegration

Dynamics 365 Apps PowerApps Power BI Apps

Demo

Create a data flow

Power BI information hierarchy

Source: Amir Netz: Power BI dataflows Whitepaper (https://go.microsoft.com/fwlink/?linkid=2034388&clcid=0x409)

✓

✓

Demo

Consume dataflow entities

(=create a dataset)

Who can use PBI dataflows?

Everyone!

https://docs.microsoft.com/en-us/power-bi/service-dataflows-overview#dataflow-capabilities-on-power-bi-premium

Behind the Scenes

Behind the Scenes

https://docs.microsoft.com/en-us/power-bi/service-dataflows-overview

Dataflows and the Data Lake

Source: Amir Netz: Power BI dataflows Whitepaper (https://go.microsoft.com/fwlink/?linkid=2034388&clcid=0x409)

“BYOSA” – Bring Your Own Storage Account

CDM Folders

https://docs.microsoft.com/en-us/power-bi/service-dataflows-azure-data-lake-integration

Dataflows Administration

Demo

Tenant settings

BYOSA

Dataflow settings

Schedule refresh

Premium – Capacity Settings

Enable dataflows

BYOSA – Use your own Data Lake

Dataflow list

Schedule Refresh

PBI Premium – Capacity Settings

Resources

Resources

• Dataflows Whitepaper (by Amir Netz) • https://go.microsoft.com/fwlink/?linkid=2011419&clcid=0x409

• Documentation• https://docs.microsoft.com/en-us/power-bi/service-dataflows-overview

• Matthew Roche• https://ssbipolar.com/2018/10/23/dataflows-in-power-bi/

• Guy in a Cube• https://www.youtube.com/channel/UCFp1vaKzpfvoGai0vE5VJ0w

• Rezas Blog• http://radacad.com/blog

• My blog (#24DaysPowerPlatform)• https://workingondata.wordpress.com/2018/12/22/24-days-of-powerplatform-day-22-what-are-

power-bi-dataflows/

WolfgangStrasser

wolfgang @ powerofbi.at

@wstrasser

https://workingondata.wordpress.com

Thanks for joining!